Systems and methods are disclosed for establishing a multi-dimensional fraud detection system and payment analysis. One method includes: receiving transaction history of a user, the transaction history including a first payment vehicle and a second payment vehicle; determining, of the received transaction history, one or more instances of switching from one the first payment vehicle to the second payment vehicle; and determining a user-specific abandonment score for the user, based on the determined instances of switching from the first payment vehicle to the second payment vehicle.
Legal claims defining the scope of protection. Each claim is shown in both the original legal language and a plain English translation.
1. A computer-implemented method of establishing a multi-dimensional fraud detection system and payment analysis, the method comprising: receiving transaction history of a user, the transaction history including a first payment vehicle and a second payment vehicle; generating a unique hash and a fraud analysis profile request upon determining a fraud analysis profile is not available, and sending the fraud analysis profile request including the unique hash to generate the fraud analysis profile from the received transaction history of the user; determining, of the received transaction history, one or more instances of switching from the first payment vehicle to the second payment vehicle; and determining a user-specific abandonment score for the user, based on the determined instances of switching from the first payment vehicle to the second payment vehicle.
This invention relates to fraud detection and payment analysis in financial transactions. The system addresses the challenge of identifying fraudulent behavior by analyzing user transaction patterns, particularly focusing on payment vehicle switching. When a user's transaction history is received, it includes data from at least two payment methods (e.g., credit cards, bank accounts). If no existing fraud analysis profile is available, the system generates a unique hash and creates a fraud analysis profile request. This request, containing the hash, is used to build a fraud analysis profile from the transaction history. The system then identifies instances where the user switches from one payment method to another within the transaction history. Based on these switching events, a user-specific abandonment score is calculated, which quantifies the likelihood of fraudulent activity. This score helps assess the risk associated with the user's payment behavior, enabling more accurate fraud detection. The method leverages multi-dimensional analysis to enhance fraud prevention by detecting patterns that may indicate suspicious or fraudulent transactions.
2. The computer-implemented method of claim 1 , further comprising: determining first billing information associated with the first payment vehicle; determining second billing information associated with the second payment vehicle; and decreasing a fraud risk score if the first billing information matches the second billing information, or increasing a fraud risk score if the first billing information does not match the second billing information.
This invention relates to fraud detection in payment processing systems, specifically methods for assessing fraud risk based on billing information consistency across multiple payment vehicles. The problem addressed is the difficulty in detecting fraudulent transactions when attackers use different payment methods, such as credit cards or digital wallets, that may not share obvious links. The solution involves analyzing billing information associated with each payment vehicle to identify discrepancies that could indicate fraud. The method involves determining billing information for a first payment vehicle and a second payment vehicle used in a transaction. Billing information may include details such as the billing address, name, or other identifiers linked to the payment method. The system then compares the billing information of the two payment vehicles. If the billing information matches, the fraud risk score is decreased, as consistency suggests legitimacy. Conversely, if the billing information does not match, the fraud risk score is increased, as discrepancies may indicate fraudulent activity. This approach helps identify potential fraud by leveraging cross-payment vehicle data consistency as a risk indicator. The method can be integrated into existing fraud detection systems to enhance accuracy and reduce false positives.
3. The computer-implemented method of claim 2 , further comprising: prompting approval of a payment transaction based on the user-specific abandonment score and the fraud risk score.
This invention relates to fraud detection and payment transaction approval in digital systems. The method addresses the problem of accurately assessing fraud risk while minimizing false positives that may lead to legitimate transactions being rejected. The system calculates a user-specific abandonment score based on user behavior patterns, such as navigation history, time spent on pages, and interaction frequency, to predict the likelihood of transaction abandonment. Additionally, a fraud risk score is generated using traditional fraud detection techniques, such as analyzing transaction data, device fingerprints, and historical fraud patterns. The method then combines these scores to determine whether to prompt approval of a payment transaction. By integrating behavioral abandonment metrics with fraud risk assessment, the system improves decision-making in transaction processing, reducing both fraud losses and unnecessary transaction declines. The approach is particularly useful in e-commerce, banking, and digital payment platforms where balancing fraud prevention and user experience is critical. The method ensures that transactions are only flagged for approval when both the abandonment and fraud risk scores meet predefined thresholds, enhancing overall system reliability.
4. The computer-implemented method of claim 2 , further comprising: computing the fraud risk score based on billing data, device data, merchant data, or timing data of the received transaction history.
This invention relates to fraud detection in financial transactions, specifically a computer-implemented method for assessing fraud risk by analyzing transaction history data. The method addresses the challenge of accurately identifying fraudulent transactions in real-time by leveraging multiple data sources to compute a fraud risk score. The core process involves receiving transaction history data, which includes details such as transaction amounts, timestamps, and involved parties. The method then processes this data to extract relevant features, such as billing information, device identifiers, merchant details, and transaction timing patterns. These features are used to compute a fraud risk score, which quantifies the likelihood that a transaction is fraudulent. The method may also involve comparing the computed score against predefined thresholds to determine whether a transaction should be flagged for further review or automatically blocked. By integrating diverse data points, the system improves fraud detection accuracy and reduces false positives compared to traditional methods that rely on limited datasets. The approach is particularly useful for financial institutions, payment processors, and e-commerce platforms seeking to enhance security while maintaining smooth transaction processing.
5. The computer-implemented method of claim 1 , further comprising: identifying, of the transaction history, one or more authorization requests associated with the first payment vehicle; determining, of the identified authorization requests, transaction data associated with the identified authorization requests, wherein the transaction data corresponds to a second payment vehicle; and determining the one or more instances of switching from the first payment vehicle to the second payment vehicle, based on the transaction data.
This invention relates to payment processing systems that analyze transaction histories to detect instances where a user switches between different payment vehicles (e.g., credit cards, debit cards, or digital wallets) during a transaction. The problem addressed is the lack of visibility into payment vehicle switching, which can lead to inefficiencies in fraud detection, transaction routing, and user experience optimization. The method involves processing a transaction history to identify authorization requests linked to a first payment vehicle. For each identified request, the system extracts transaction data, which includes details about the payment vehicle used to complete the transaction. By comparing the first payment vehicle with the actual payment vehicle used (the second payment vehicle), the system determines instances where the user switched payment methods. This analysis helps financial institutions and payment processors detect patterns, improve fraud prevention, and optimize transaction routing by understanding how users interact with multiple payment options. The solution enhances transparency in payment processing and enables better decision-making based on real transaction behavior.
6. The computer-implemented method of claim 1 , further comprising: receiving a first payment authorization request associated with the user; receiving a fraud risk score associated with the first payment authorization request; receiving an institutional risk tolerance score; and prompting approval of the first payment authorization request based on the user-specific abandonment score, the fraud risk score, and the institutional risk tolerance score.
This invention relates to fraud detection and risk management in payment processing systems. The method involves assessing the likelihood of a user abandoning a transaction based on user-specific behavioral data, such as browsing patterns, purchase history, and interaction timelines. A user-specific abandonment score is generated to predict whether the user will complete the transaction or leave the process. The method further evaluates fraud risk by analyzing transaction details, such as payment method, amount, and location, to produce a fraud risk score. Additionally, an institutional risk tolerance score is used to determine the acceptable level of risk for the organization processing the payment. The system combines these three scores—the user-specific abandonment score, the fraud risk score, and the institutional risk tolerance score—to determine whether to approve or reject a payment authorization request. This approach enhances decision-making by balancing user behavior, fraud risk, and organizational risk preferences, reducing false positives and improving transaction approval rates while mitigating fraud. The method is particularly useful in e-commerce, banking, and financial services where fraud prevention and user experience optimization are critical.
7. The computer-implemented method of claim 6 , further comprising: denying the first payment authorization request unless the fraud risk score is less than the institutional risk tolerance score.
This invention relates to fraud detection in payment processing systems. The problem addressed is the need to prevent fraudulent transactions while minimizing false positives that could disrupt legitimate transactions. The method involves analyzing a payment authorization request to determine a fraud risk score, which quantifies the likelihood that the transaction is fraudulent. This score is compared to an institutional risk tolerance score, which represents the acceptable level of risk for the financial institution. If the fraud risk score exceeds the institutional risk tolerance score, the payment authorization request is denied. The method ensures that only transactions deemed sufficiently low-risk are approved, reducing fraud losses while maintaining operational efficiency. The fraud risk score may be derived from various factors, including transaction history, user behavior patterns, and external threat intelligence. The institutional risk tolerance score can be dynamically adjusted based on changing risk conditions or business policies. This approach provides a flexible and adaptive fraud detection mechanism that balances security and usability.
8. The computer-implemented method of claim 6 , further comprising: receiving a profile associated with the user; computing an initial fraud risk score based on the profile and transaction data of the first payment authorization request; receiving a second payment authorization request; and updating initial fraud risk score based on the second payment authorization request.
This invention relates to fraud detection in payment processing systems. The problem addressed is the need for dynamic fraud risk assessment that adapts to new transaction data in real-time. Traditional fraud detection systems often rely on static profiles or limited transaction history, which can fail to detect evolving fraud patterns. The method involves receiving a user profile containing historical data and transaction patterns. An initial fraud risk score is computed by analyzing this profile alongside transaction data from a first payment authorization request. The system then receives a second payment authorization request and updates the initial fraud risk score based on this new transaction data. This dynamic updating allows the system to continuously refine its fraud assessment as new transactions occur, improving accuracy over time. The method may also include generating a fraud alert if the updated risk score exceeds a predefined threshold, enabling real-time intervention. The system can further adjust the risk score based on additional factors such as transaction frequency, location, or device information. By continuously updating the risk score with each new transaction, the system provides a more responsive and adaptive fraud detection mechanism compared to static scoring models. This approach helps financial institutions reduce false positives and improve fraud prevention.
9. The computer-implemented method of claim 1 , wherein the transaction history includes, one or more of: an identifier of the merchant; an identifier of the user; an identifier of one or more payment means accessed for payment of the transaction; an itemization of the goods and/or services transacted for; any geographical and/or temporal information of the transaction; any taxes, tips, and/or gratuities; any discounts, coupons, reductions; any fees directed towards acquirers, issuers, payment networks; currency exchange rates; and any other attributes of the payment transaction.
This invention relates to a computer-implemented method for processing and analyzing transaction history data in payment systems. The method addresses the need for comprehensive transaction record-keeping, enabling detailed tracking and analysis of financial transactions. The transaction history includes a wide range of attributes to provide a complete record of each transaction. These attributes may include an identifier for the merchant involved, an identifier for the user making the payment, and identifiers for one or more payment methods used to complete the transaction. The method also captures an itemized breakdown of the goods or services purchased, along with any geographical or temporal details of the transaction. Additional financial details such as taxes, tips, gratuities, discounts, coupons, reductions, and fees paid to acquirers, issuers, or payment networks are recorded. Currency exchange rates and other relevant transaction attributes are also included. This detailed transaction history allows for enhanced financial tracking, fraud detection, and analytical capabilities in payment processing systems.
10. The computer-implemented method of claim 1 , further comprising: prompting approval of a payment transaction based on the user-specific abandonment score.
A computer-implemented method addresses the problem of payment transaction abandonment in digital commerce by analyzing user behavior to predict and mitigate potential transaction failures. The method involves generating a user-specific abandonment score, which quantifies the likelihood that a user will abandon a payment transaction before completion. This score is derived from historical transaction data, user interaction patterns, and contextual factors such as device type, network conditions, or time of day. The method further includes prompting approval of a payment transaction based on the abandonment score, allowing for proactive intervention to reduce transaction drop-offs. For example, if the score indicates a high abandonment risk, the system may trigger additional user prompts, simplified checkout steps, or alternative payment options to encourage completion. The method may also integrate with fraud detection systems to balance risk assessment with user experience. By dynamically adjusting transaction flows based on abandonment risk, the system aims to improve conversion rates and reduce revenue loss from incomplete transactions. The approach leverages machine learning or statistical models to refine abandonment predictions over time, adapting to evolving user behaviors and market trends.
11. A decentralized computer system for establishing a multi-dimensional fraud detection system and payment analysis, the system comprising: a data storage device storing instructions for establishing a multi-dimensional fraud detection system and payment analysis; and a processor configured to execute the instructions to perform a method including: receiving transaction history of a user, the transaction history including a first payment vehicle and a second payment vehicle; generating a unique hash and a fraud analysis profile request upon determining a fraud analysis profile is not available, and sending the fraud analysis profile request including the unique hash to generate the fraud analysis profile from the received transaction history of the user; determining, of the received transaction history, one or more instances of switching from the first payment vehicle to the second payment vehicle; and determining a user-specific abandonment score for the user, based on the determined instances of switching from the first payment vehicle to the second payment vehicle.
A decentralized computer system detects fraud and analyzes payments by monitoring user transaction histories across multiple payment methods. The system addresses the challenge of identifying fraudulent behavior in decentralized environments where traditional centralized fraud detection methods are ineffective. The system stores transaction data and processes it to detect patterns indicative of fraud, particularly focusing on payment vehicle switching behavior. The system receives a user's transaction history, which includes records of payments made using different payment methods (e.g., credit cards, digital wallets). If no existing fraud analysis profile is available for the user, the system generates a unique hash and creates a fraud analysis profile request. This request, containing the hash, is used to generate a fraud analysis profile based on the transaction history. The system then analyzes the transaction data to identify instances where the user switches from one payment method to another. Based on these switching events, the system calculates a user-specific abandonment score, which quantifies the likelihood of fraudulent activity. This score helps assess the risk associated with the user's payment behavior, enabling more accurate fraud detection in decentralized systems. The system's multi-dimensional approach improves fraud detection by considering multiple payment methods and behavioral patterns.
12. The system of claim 11 , wherein the processor is further configured for: determining first billing information associated with the first payment vehicle; determining second billing information associated with the second payment vehicle; and decreasing a fraud risk score if the first billing information matches the second billing information, or increasing a fraud risk score if the first billing information does not match the second billing information.
A payment processing system analyzes transaction data to assess fraud risk by comparing billing information from multiple payment vehicles. The system processes transactions involving at least two payment vehicles, such as credit cards or digital wallets, linked to a user account. For each transaction, the system retrieves billing information associated with each payment vehicle, including details like billing addresses, names, or other identifiers. The system then compares the billing information from the first payment vehicle to the second payment vehicle. If the billing information matches, the system decreases a fraud risk score, indicating a lower likelihood of fraudulent activity. Conversely, if the billing information does not match, the system increases the fraud risk score, signaling a higher potential for fraud. This comparison helps detect inconsistencies that may indicate unauthorized or fraudulent use of payment vehicles. The system may also use the fraud risk score to determine whether to approve, reject, or flag a transaction for further review. This approach enhances security by leveraging cross-payment vehicle data to identify discrepancies that traditional fraud detection methods might miss.
13. The system of claim 12 , wherein the processor is further configured for: prompting approval of a payment transaction based on the user-specific abandonment score and the fraud risk score.
A system for payment transaction approval evaluates user behavior and fraud risk to determine whether to prompt approval of a transaction. The system monitors user interactions with a payment interface to detect abandonment behavior, such as hesitation or incomplete actions, and generates a user-specific abandonment score based on these interactions. Additionally, the system assesses transaction data to calculate a fraud risk score, which quantifies the likelihood of fraudulent activity. The processor then uses both the abandonment score and the fraud risk score to decide whether to prompt for approval of the payment transaction. This approach helps balance user experience with fraud prevention by considering both behavioral and transactional risk factors. The system may also include a user interface for displaying transaction details and a communication module for transmitting approval requests to a user device. The abandonment score is derived from metrics like interaction time, input patterns, and session duration, while the fraud risk score incorporates factors such as transaction amount, location, and historical fraud data. By integrating these scores, the system dynamically adjusts approval prompts to reduce false declines while mitigating fraud risks.
14. The system of claim 12 , wherein the processor is further configured for: computing the fraud risk score based on billing data, device data, merchant data, or timing data of the received transaction history.
This invention relates to fraud detection systems for transaction processing. The system analyzes transaction history to assess fraud risk by evaluating multiple data sources. The processor computes a fraud risk score using billing data, device data, merchant data, or timing data from the transaction history. Billing data may include payment details, recipient information, or transaction amounts. Device data may involve device identifiers, geolocation, or network characteristics. Merchant data could encompass merchant reputation, transaction patterns, or industry-specific risk factors. Timing data may analyze transaction frequency, time of day, or historical behavior deviations. The system processes this data to identify anomalies or suspicious patterns indicative of fraudulent activity. The fraud risk score quantifies the likelihood of fraud, enabling automated decision-making or flagging high-risk transactions for further review. This approach enhances fraud detection accuracy by leveraging diverse data points to detect sophisticated fraud schemes. The system may integrate with payment networks, financial institutions, or merchant platforms to monitor transactions in real-time. By analyzing multiple data dimensions, the system improves fraud prevention while minimizing false positives.
15. The system of claim 11 , wherein the processor is further configured for: identifying, of the transaction history, one or more authorization requests associated with the first payment vehicle; determining, of the identified authorization requests, transaction data associated with the identified authorization requests, wherein the transaction data corresponds to a second payment vehicle; and determining the one or more instances of switching from the first payment vehicle to the second payment vehicle, based on the transaction data.
A payment processing system analyzes transaction histories to detect instances where a user switches between payment methods. The system identifies authorization requests linked to a primary payment vehicle, such as a credit card, and extracts transaction data from these requests. The transaction data reveals instances where a secondary payment vehicle, such as a debit card or digital wallet, was used instead of the primary one. By comparing the transaction histories, the system determines when and how often the user switches between the primary and secondary payment methods. This functionality helps financial institutions or payment processors track payment method preferences, optimize transaction routing, or detect potential fraud by identifying unusual switching patterns. The system processes transaction records in real-time or batch mode, ensuring accurate detection of payment method transitions. The analysis may include metadata such as transaction timestamps, merchant identifiers, and payment method details to ensure precise identification of switching events. This capability enhances payment flexibility and security by providing insights into user behavior and payment method usage trends.
16. The system of claim 11 , wherein the processor is further configured for: receiving a first payment authorization request associated with the user; receiving a fraud risk score associated with the first payment authorization request; receiving an institutional risk tolerance score; and prompting approval of the first payment authorization request based on the user-specific abandonment score, the fraud risk score, and the institutional risk tolerance score.
This invention relates to a fraud detection and risk management system for payment processing. The system addresses the problem of balancing fraud prevention with user experience by dynamically assessing risk factors to determine whether to approve or reject payment transactions. The system includes a processor configured to generate a user-specific abandonment score, which predicts the likelihood of a user abandoning a transaction based on historical behavior and contextual data. The processor also evaluates fraud risk scores associated with payment authorization requests and institutional risk tolerance scores, which define the acceptable level of risk for the financial institution. By integrating these three factors—the user-specific abandonment score, the fraud risk score, and the institutional risk tolerance score—the system determines whether to approve a payment request. This approach minimizes false positives in fraud detection while maintaining security, ensuring that transactions are only rejected when the combined risk exceeds the institution's tolerance. The system dynamically adjusts approval decisions based on real-time risk assessments, improving both fraud prevention and user satisfaction.
17. The system of claim 11 , wherein the transaction data includes, one or more of: an identifier of the merchant; an identifier of the user; an identifier of one or more payment means accessed for payment of the transaction; an itemization of the goods and/or services transacted for; any geographical and/or temporal information of the transaction; any taxes, tips, and/or gratuities; any discounts, coupons, reductions; any fees directed towards acquirers, issuers, payment networks; currency exchange rates; and any other attributes of the payment transaction.
The invention relates to a system for processing and analyzing transaction data in payment networks. The problem addressed is the lack of comprehensive transaction data collection and utilization, which limits the ability to optimize payment processing, fraud detection, and financial analytics. The system captures detailed transaction information to enhance decision-making and improve transaction handling. The system includes transaction data that encompasses various attributes of a payment transaction. This data includes identifiers for the merchant and user involved, details of the payment means used (such as credit cards, digital wallets, or bank transfers), and an itemized breakdown of goods or services purchased. Additionally, the system records geographical and temporal details of the transaction, including location and time, as well as financial aspects like taxes, tips, gratuities, discounts, coupons, and reductions. It also tracks fees associated with acquirers, issuers, and payment networks, along with currency exchange rates and other relevant transaction attributes. This comprehensive data collection enables advanced analytics, fraud detection, and personalized financial services. The system ensures that all transaction details are captured and processed efficiently, improving transparency and accuracy in payment transactions.
18. The system of claim 11 , wherein the processor is further configured for: prompting approval of a payment transaction based on the user-specific abandonment score.
A system for managing payment transactions includes a processor that evaluates user behavior to determine a user-specific abandonment score, which quantifies the likelihood of a user abandoning a transaction before completion. The system monitors user interactions with a payment interface, such as navigation patterns, time spent on steps, and input errors, to generate this score. The processor then uses the score to assess transaction risk and prompts approval of the payment transaction based on the abandonment score. If the score indicates a high likelihood of abandonment, the system may intervene with prompts, discounts, or simplified steps to encourage completion. The system may also integrate with fraud detection mechanisms to cross-reference abandonment behavior with known fraud patterns. The processor dynamically adjusts the approval criteria based on historical data and real-time user behavior to optimize transaction success rates while minimizing fraud risk. The system is designed for e-commerce platforms, digital wallets, and other payment processing environments where user abandonment leads to lost revenue. The abandonment score is recalculated in real-time as the user progresses through the payment flow, allowing for adaptive decision-making. The system may also log and analyze abandonment patterns to refine future transaction handling strategies.
19. A non-transitory machine-readable medium storing instructions that, when executed by a server, cause the server to perform a method for establishing a multi-dimensional fraud detection system and payment analysis, the method including: receiving transaction history of a user, the transaction history including a first payment vehicle and a second payment vehicle; generating a unique hash and a fraud analysis profile request upon determining a fraud analysis profile is not available, and sending the fraud analysis profile request including the unique hash to generate the fraud analysis profile from the received transaction history of the user; determining, of the received transaction history, one or more instances of switching from the first payment vehicle to the second payment vehicle; and determining a user-specific abandonment score for the user, based on the determined instances of switching from the first payment vehicle to the second payment vehicle.
This invention relates to fraud detection and payment analysis in financial transactions. The system addresses the challenge of identifying fraudulent behavior by analyzing user transaction patterns, particularly instances where users switch between different payment methods. The solution involves creating a multi-dimensional fraud detection framework that evaluates transaction histories to detect suspicious activity. The system receives a user's transaction history, which includes records of payments made using at least two different payment vehicles (e.g., credit cards, bank accounts, digital wallets). If no existing fraud analysis profile is available for the user, the system generates a unique hash and a fraud analysis profile request. This request is used to create a fraud analysis profile based on the transaction history. The system then analyzes the transaction history to identify instances where the user switches from one payment vehicle to another. These switching events are used to calculate a user-specific abandonment score, which quantifies the likelihood of fraudulent behavior based on payment method changes. The abandonment score helps assess whether the user's transaction patterns indicate potential fraud, enabling more accurate fraud detection and risk assessment. The system operates on a server and relies on machine-readable instructions to perform these functions.
20. The machine readable medium of claim 19 , wherein the method further comprises: determining first billing information associated with the first payment vehicle; determining second billing information associated with the second payment vehicle; and decreasing a fraud risk score if the first billing information matches the second billing information, or increasing a fraud risk score if the first billing information does not match the second billing information.
This invention relates to fraud detection in payment processing systems, specifically for reducing fraud risk by analyzing billing information consistency across multiple payment vehicles. The system compares billing details, such as addresses or names, from different payment methods (e.g., credit cards, bank accounts) linked to a user or transaction. If the billing information matches, the system lowers the fraud risk score, indicating higher legitimacy. If the billing information does not match, the system raises the fraud risk score, flagging potential fraud. The fraud risk score is used to assess transaction validity, helping financial institutions or payment processors identify suspicious activity. This method enhances security by cross-referencing payment vehicle data to detect inconsistencies that may indicate fraudulent use. The system may be implemented in software or hardware components that process payment transactions and evaluate billing information for discrepancies. The invention improves fraud detection accuracy by leveraging billing information as a verification factor, reducing false positives and negatives in fraud detection systems.
Cooperative Patent Classification codes for this invention. Click any code to explore related patents in that topic.
December 27, 2018
March 8, 2022
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.